Unclassified Extracts from ARPA Semiannual Technical Note - Application of Recognition Theory to Detection, Discrimination, and Decision

Abstract

The application of pattern recognition to missile detection and decoy discrimination was investigated. As a result of these investigations, a mathematical theory was developed to achieve the optimum combination of measurable or computable target properties to achieve the separation of target types, based on their observable properties. The input to the theory is the set of discrimination techniques and useful target-signature parameters. The theory operates on these inputs to determine their optimum method of combination and thus determine what the essential properties of targets are that allow their recognition. An important attribute of the theory is that it does not become obsolete as new techniques are developed and breakthroughs are achieved in methods of distinguishing one type of target or threat from another. The achievement of the present theory is that it states how techniques developed by others are to be combined.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 30, 1960
Accession Number
AD0257522

Entities

Organizations

  • Melpar

Tags

Communities of Interest

  • C4I
  • Human Systems
  • Sensors
  • Space
  • Weapons Technologies

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Automata
  • Character Recognition
  • Computational Science
  • Computer Programs
  • Computers
  • Coordinate Systems
  • Detection
  • Eigenvectors
  • Fish
  • Information Science
  • Information Theory
  • New York
  • Pattern Recognition
  • Probability
  • Supervised Machine Learning
  • Three Dimensional

Readers

  • Sensor Fusion and Tracking Systems.
  • Systems Analysis and Design
  • Theoretical Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms